Sweet. I've found one here in the States for $78 shipped...I've sent HF Signals a message asking if there was going to be a 2 month delay in shipping....hope to hear back soon or I'm buying the local
|
It was up to Monsieur Trager to do all the searching of the forum -? as far as I was concerned I was in the Cone? of Silence.? I get the feeling though that I was the only one who gave half sensible answers to is queries a while back.? (many by PM).
"On my Bitx40 from Dec of 2016, D15 and D16 do not have zero ohm resistors, they are left empty..
The board accommodates either a Sot23-3 for the Bat54s, or a fallback of the dual 1n4148's plus pot."
you will by now have seen the photo? which shows SM resistors or SOMETHING mounted at the? D15 and 16 board that Trager was working on (did he put them in ?? perhaps yes to add in the resistance taken out when the? 100R trim pot was removed ) so perhaps this was an interim revision of the copper ?? or a Stuff UP.?
He did not buy the boards; Someone did some time back when I do not know. The board SN is on the photo but I would suggest for instance AF and such interested parties? stay out of it because Trager does repairs for money it seems and he has done a lot of work he cannot bill for and will have a temper to match I would guess.
I quote -?
"best bet is to upgrade to Allard's v2 firmware and use clk0 from the si5351."? Appreciate the point but? ?Trager and I took it as a given that this board was ready to run and did not require an upgrade c/off? ? Santa Claus , the Archangel Gabriel or anyone else.? ?It was essentially a DOA PCB,? and there were four more with it for rework.? He does not know if they all had same issue or the purchaser tested one and found it faulty and therefore wanted them all looked at.??
|
Re: the cause for the spurs, found!
I had to re-bias after swapping the TX transistors (Q20, Q21 and Q22) in June. I wondered why TX-power on SSB went down - and the reason was that Q20 was not on the correct operating point. Allisson gave the right hint: At ideal bias the collector of Q20 (if a BFR106 is used) shall have about 6V. And just by looking up that one, I am wondering now if a lot of solutions (or the right approaches) have not already been discussed in the past.
|
I believe the 12mhz crystals for the Bitx40 are all from the same bin after sorting by frequency.
The filter uses them at series resonance, the BFO uses the crystal at the parallel resonant frequency.
This gives the correct offset for the Bitx40.
Trimmer cap not needed.
If you do want to move the BFO, best bet is to upgrade to Allard's v2 firmware and use clk0 from the si5351.
Am curious how much higher "slightly high" is.
Not that it matters now.
I'm not sure how thorough the check at HFSignals was, but doubt it was deaf when shipped.
This sort of thing has typically been something like the broken toroid wire or cold solder joint
previously mentioned, since they do check boards before shipping.
Farhan's video? ??
shows how to go through the receiver and figure out exactly where it goes deaf.
He just touches a 1 meter piece of wire (an antenna that picks up local QRM) to the?
various stages, expecting the resultant noise to get louder and louder as he progresses
back from the audio amp to the RF front end.?
That technique would likely have pinpointed the reason for the deaf receiver.
>? what if the factory installed by mistake at D15 and D16 posns not zero ohm resistors as they should be ,? but? even DIODES ?
On my Bitx40 from Dec of 2016, D15 and D16 do not have zero ohm resistors, they are left empty..
The board accommodates either a Sot23-3 for the Bat54s, or a fallback of the dual 1n4148's plus pot.
Traces from the Sot23 go around D15,D16 and tie off directly to T4.
A quick query to the forum and we could have told you of the few differences between board and schematic
(as enumerated in my previous post), that got posted here a number of times in 2017.
FYI, I believe the boards get auto-stuffed for the surface mount parts.
So most stuffing errors would happen across a thousand boards or so due to a wrong reel of parts.
And we'd hear all about it in the forum if it wasn't caught at HFSignals.??
Never been an issue.
OK, I'm over it too.
Except to say I suspect that this may have been a pretty simple problem?
before?all the shotgunning happened.
Jerry?
toggle quoted message
Show quoted text
On Tue, Nov 6, 2018 at 03:36 PM, Timothy Fidler wrote:
He matched the xtals but did not measure the passband as per your above.? I warned him that he would? be fitting xtals of unknown motional params to? a set of capacitors that were not matched.? He did not have the equipment that could do more.? Bear in mind the thing was utterly useless , utterly deaf before. Afterwards it could Rx SSB.? So that's progress. What he DID find with the as removed xtal set was that one which was? slightly high (and therefore should be the one at the BFO posn for LSB selection)? was IN THE FILTER SET . and the one at the BFO posn was at nominal frequency - I have to trust him that he marked them properly before removal and test.? So; no trimmer capacitor, regardless of as published schematic? in posn and a lousy job of selection of filter xtals. Perhaps? someone hungover on a Monday ?? about that time he found he had plenty of carrier? bleed through and also that the BM was not as per the schematic on the desk? (I am not sure if this was measured at PA out but I suspect so )? he decided to give up - it was? a job he was doing for someone else.? ?TEF
|
my recent ubitx was delivered with the LSB BFO way off, but the USB side okay.? try both USB and LSB on appropriate bands.? when BFO is right, the SSB and CW audio are quite nice.? I am surprised the tune-up instructions don't provide a target value for these BFO settings, as once you are close it is easy to adjust.?
73 Curt
|
Built this circuit up, and caught the need to use a 560 ohm resistor to feed the transistor that drives the LED.? Using a orange LED and a LDR with 20k dark resistance.? Only on very high signals do I get illumination, and even then no perceived AGC action.? Strangely then when I increase gain control, the LED gets brighter which implies awful poor isolation.? To install this thing I disconnected the wire from the hot (or upper) end of the audio gain control - attached this wire to the AGC circuit input and the output onto this pot.? I connected circuit ground to the ground end of the pot, and biased it from the DC connector.?
Circuit seems to work in the videos but not in my build.? Thought about tweaking parts on the input amplifier stage, but as light doesn't lead to attenuation wondering if my build has 2 flaws.? Anyone around who had success with this?? I am not seeking a robust AGC, but somethng to squash the loudest signals, so I need to use it less.? If this thing worked I would implement an AGC on/off switch.?
73 Curt
|
Finally got my ubitx working mostly correctly. Everything is default, firmware is 2.0. After following tuneup procedures, I'm able to tune to local AM stations and hear properly, audio is great aside from the unignorable louuud tone. It makes listening with headphones unbearable, but I can still hear the (strong) signal under it. The pitch of the tone does change with bfo adjustment and recalibration, but I can't change those as the quality of the actual signal is hurt. The tone is also significantly quieter, although still present on USB in comparison to LSB. Tuning does NOT change the pitch of the tone.
Could this be a tuneup error? Or something deeper?
Thanks!?
|
 here it is after? a lot of rework with the dual diode pkg as per factory installation . The SX xtal was thrown out an a full HC49U one (Digikey) was put in from the new matched set. As you can see, it has had? a bit of rework.? NOW here is a Dorothy Dixer... what if the factory installed by mistake at D15 and D16 posns not zero ohm resistors as they should be ,? but? even DIODES ? (this was not tested for because he was sick of working on the PCB by this stage )? or perhaps two unmatched resistors.. that would really louse it up . Perhaps a few,? as in a production batch? were issued in this state and the other buyers have just given up ?? Anyways I get nothing to take away from this discussion so that's all I know and all he knows and frankly;? I am "Over it "? ? TEF
|
He matched the xtals but did not measure the passband as per your above.? I warned him that he would? be fitting xtals of unknown motional params to? a set of capacitors that were not matched.? He did not have the equipment that could do more.? Bear in mind the thing was utterly useless , utterly deaf before. Afterwards it could Rx SSB.? So that's progress. What he DID find with the as removed xtal set was that one which was? slightly high (and therefore should be the one at the BFO posn for LSB selection)? was IN THE FILTER SET . and the one at the BFO posn was at nominal frequency - I have to trust him that he marked them properly before removal and test.? So; no trimmer capacitor, regardless of as published schematic? in posn and a lousy job of selection of filter xtals. Perhaps? someone hungover on a Monday ?? about that time he found he had plenty of carrier? bleed through and also that the BM was not as per the schematic on the desk? (I am not sure if this was measured at PA out but I suspect so )? he decided to give up - it was? a job he was doing for someone else.? ?TEF
|
Hmm.
Arv asked?
>? But?you?also made some claims that need backing up.?
>? ? ? HF Signals back ordering...?? ??Is that really true?? Where did you get that information?
JGaffke said that the old style balanced modulator had two discrete diodes and a pot.
Which I suspect Arv knows full well because he has built a few Bitx's like that.
Old and new balanced modulators both work fine.
The new one is arguably better because it has one less knob to set wrong,
and the two diodes on one die are well enough matched.
Nobody has gotten a Bitx40 with discrete diodes and pot over the last couple years.
And very few have been tripped up by that.
But yes, I'll agree that schematics up on HFSignals should be kept up to date.
And old versions of the schematics should remain available.
Jerry
toggle quoted message
Show quoted text
On Tue, Nov 6, 2018 at 03:20 PM, Timothy Fidler wrote:
It has been backed up by J Gaffke on this BBS already (that is the issue of the BM change)
The board as photographed (not mine) shows the old style BM with two diodes , a so called half bridge BM with a balancing potentiometer.? ?According to to M Trager this is NOT what he received.? ?THe "Better" BM seems to be of the style of the uBitx design. M T did send me some photos of his reworked board but he concentrated on the Cohn filter area. Will look. NB the missing trimmer capacitor above makes it very difficult to set up the board if the crystal selection at the factory is not perfect.?
|
Ah, I do vaguely remember that stuff from long long ago in a thread far away ...
Yup, there's a few spots where the currently shipped Bitx40 does not agree with the schematic.
That balanced modulator pot and diodes, the trimmer cap at the BFO crystal, and the analog VFO coil.
That's it.? Nothing that gave me more than a moment's pause when I first ran into it.
Did he measure the passband or was he just shotgunning when he swapped out all those crystals?
Swapping out crystals would not be a good idea without careful measurements of both what you have
and what you are putting in.? Could be something as simple as a cold solder joint.
You say he measured all the incoming crystals to match for frequency, so that at least sounds good.
Presented with a rig that simply did not receive, I might try jumping across that filter to verify it then
received ok, then the add back in the existing crystals a few at a time.
There are several possible causes of carrier leakage.
Yes, a fouled up crystal filter could be partly responsible.
Board layout and lack of shielding can contribute.
Loose wires lying about where they shouldn't, such as
clk2 from the Raduino to the Bitx40 board passing over the BFO.
And if using clk0 for the BFO (not the Bitx40's stock crystal oscillator),
there's crosstalk within the si5351.
Some residual carrier is not a deal breaker for me, or for the regulatory authorities.
The Bitx40 is fairly clean with respect to the harmonics and spurs that currently bedevil the uBitx.
But if the crystal filter truly truly fouled up on arrival, that might call for a sip or two.
There were some reports here of failed Bitx40's?
Usually it was at one of the toroids, a wire broken in transit or a cold solder joint that happened
to work ok during checkout at HFSignals.? Occasionally, the Bitx40 arrived looking like a gorilla
had thrown it against the wall, better packaging has mostly done away with that issue.
Most of the few hams that checked in here with a bad Bitx40 managed to get them going?
Jerry?
toggle quoted message
Show quoted text
On Tue, Nov 6, 2018 at 01:50 PM, Timothy Fidler wrote:
The Bitx40 that M Trager (lately of these pages ) rxd was a nightmare and he gave up on it.? I had him search the cct? and he related what J Gaf above has noted.? Trager had changed out the full xtal set by this time (as received and even after calib , it did not even receive)... He did have it receiving fine after a lot of work on the PCB including ordering? 20 more xtals from Digkey and matching them [time is money] .. once he and I realised the PCB as supplied was different from the schematic,? he threw up his hands and said - Genug.. I am over it.? At that stage there was a lot of carrier bleed through the filter? so he and I were suspicious of the Bal modulator and the diodes .... you could try to hunt him up on these pages but I 'd say he is trying to find solice in the bottom of a Jack Daniels bottle at the moment.
|
It has been backed up by J Gaffke on this BBS already (that is the issue of the BM change)  The board as photographed (not mine) shows the old style BM with two diodes , a so called half bridge BM with a balancing potentiometer.? ?According to to M Trager this is NOT what he received.? ?THe "Better" BM seems to be of the style of the uBitx design. M T did send me some photos of his reworked board but he concentrated on the Cohn filter area. Will look. NB the missing trimmer capacitor above makes it very difficult to set up the board if the crystal selection at the factory is not perfect.?
|
Timothy But you also made some claims that need backing up.?
- HF Signals back ordering...?
Is that really true?? Where did you get that information??
- On Tue, Nov 6, 2018 at 12:59 PM, Timothy Fidler wrote:
HFSigs seem to have made some hardware mods to the Bitx40
Which boards does this comment apply to?? Where did you get that information?
We need reliable information.? Arv _._
|
Yes.??? I was about to put it on eBay.
I could ship it tomorrow.?? Priority or std,? it should get to about anywhere in the US within 3-4 days.
Priority 3 or less. ? ? ? ? Sent from for Windows 10 ?
toggle quoted message
Show quoted text
|
The Bitx40 that M Trager (lately of these pages ) rxd was a nightmare and he gave up on it.? I had him search the cct? and he related what J Gaf above has noted.? Trager had changed out the full xtal set by this time (as received and even after calib , it did not even receive)... He did have it receiving fine after a lot of work on the PCB including ordering? 20 more xtals from Digkey and matching them [time is money] .. once he and I realised the PCB as supplied was different from the schematic,? he threw up his hands and said - Genug.. I am over it.? At that stage there was a lot of carrier bleed through the filter? so he and I were suspicious of the Bal modulator and the diodes .... you could try to hunt him up on these pages but I 'd say he is trying to find solice in the bottom of a Jack Daniels bottle at the moment.
|
That older Bitx40 would have to be at least 2 years old, from before they had a Raduino.
The modulator then was two discrete 1n4148a's and a pot.
The Bat54s now used has the two diodes on one die and they are thus well matched,
but the internal connection in the Bat54s precludes the use of a pot.
If you want to play with discrete diodes and pot to see if you can improve
carrier rejection, that would be trivial to?hack into a current day Bitx40 or uBitx.
No need to browse hamfest flea markets for ancient Bitx's.
Crystal matching is generally pretty good on these rigs, but it's tough to take out?
a strong carrier that's just a few hundred hertz away from the passband.
With current uBitx firmware and with Allard's v2 firmware for the Bitx40,
the BFO is driven from clk0 of the si5351 and thus easily adjusted.
Jerry
toggle quoted message
Show quoted text
On Tue, Nov 6, 2018 at 12:59 PM, Timothy Fidler wrote:
B Shep ; reading between the lines on some work I did with M Trager , HFSigs seem to have made some hardware mods to the Bitx40 (modulator area and leaving out the trim cap on the BFO OSC which is a critical item if the crystal matching is not as good as it might be) and Like Kenny Rogers? ,after going to Nam, "it's not the man it used to be " .? You may well be better off buying an older unused board than more recent one. My Tuppence.
|
On Tue, Nov 6, 2018 at 12:59 PM, Timothy Fidler wrote:
HFSigs seem to have made some hardware mods to the Bitx40
Huh. Got any more details or sources? Not that I don't believe you, just...trust but verify, right? -- Jon Tabor KI7JYE
|
B Shep ; reading between the lines on some work I did with M Trager , HFSigs seem to have made some hardware mods to the Bitx40 (modulator area and leaving out the trim cap on the BFO OSC which is a critical item if the crystal matching is not as good as it might be) and Like Kenny Rogers? ,after going to Nam, "it's not the man it used to be " .? You may well be better off buying an older unused board than more recent one. My Tuppence.
|
On Tue, Nov 6, 2018 at 12:58 PM, Timothy Fidler wrote:
B Shep -? reading between the lines on some work I did with M Trager , HFSigs seem to have made some hardware mods to the Bitx40 (modulator area and leaving out the trim cap on the BFO OSC which is a critical item if the crystal matching is not as good as it might be) and Like Kenny Rogers? ,after going to Nam, "it's not the man it used to be " .? You may well be better off buying an older unused board than more recent one. My Tuppence.
|
Interesting. I bought one (my first!) last week, and it shipped yesterday.
Jon
KI7JYE
|
